Multi-UAV IRS-Assisted Communications: Multi-Node Channel Modeling and Fair Sum-Rate Optimization via Deep Reinforcement Learning

Unmanned Aerial Vehicles (UAVs) combined with Intelligent Reflective Surfaces (IRSs) represent a cutting-edge technology for improving the channel capacity of wireless communications, by capitalizing on UAVs’ 3D mobility coupled IRSs’ smart radio capabilities. This work envisions scenario in which swarm UAVs equipped IRSs serves multiple Internet Things (IoT) Ground Nodes (GNs) concurrently transmitting to single Base Station (BS) via OFDMA. The huge number passive elements composing introduces significant complexity mission design. Therefore, each IRS is divided into patches that can be simultaneously used serve different nodes. Considering general Rician fading, comprehensive model IRS-assisted UAV-aided networks derived. Then, multi-objective mixed-integer non-linear programming problem conceived maximize sum-rate GNs and, at same time, minimize difference among users’ data rates, jointly optimizing trajectories and phase shift matrices. non-convex problem, reformulated terms scheduling (i.e., patch-GN assignment), challenging solve. Hence, it rearranged as Markov Decision Process quasi-optimal solution obtained Deep Reinforcement Learning. Extensive simulation analysis performed validate results accuracy proposed model.